parse5 2.2.3 → 3.0.3

package/lib/{common → utils}/merge_options.js

@@ -1,13 +1,13 @@
-'use strict';
-
-module.exports = function mergeOptions(defaults, options) {
-    options = options || {};
-
-    return [defaults, options].reduce(function (merged, optObj) {
-        Object.keys(optObj).forEach(function (key) {
-            merged[key] = optObj[key];
-        });
-
-        return merged;
-    }, {});
-};
+'use strict';
+
+module.exports = function mergeOptions(defaults, options) {
+    options = options || Object.create(null);
+
+    return [defaults, options].reduce(function (merged, optObj) {
+        Object.keys(optObj).forEach(function (key) {
+            merged[key] = optObj[key];
+        });
+
+        return merged;
+    }, Object.create(null));
+};

package/lib/utils/mixin.js

@@ -0,0 +1,18 @@
+'use strict';
+
+var Mixin = module.exports = function (host) {
+    var originalMethods = {},
+        overriddenMethods = this._getOverriddenMethods(this, originalMethods);
+
+    Object.keys(overriddenMethods).forEach(function (key) {
+        if (typeof overriddenMethods[key] === 'function') {
+            originalMethods[key] = host[key];
+            host[key] = overriddenMethods[key];
+        }
+    });
+};
+
+Mixin.prototype._getOverriddenMethods = function () {
+    throw new Error('Not implemented');
+};
+

package/package.json

@@ -1,30 +1,25 @@
 {
   "name": "parse5",
-  "description": "WHATWG HTML5 specification-compliant, fast and ready for production HTML parsing/serialization toolset for Node.js",
-  "version": "2.2.3",
+  "description": "HTML parsing/serialization toolset for Node.js. WHATWG HTML Living Standard (aka HTML5)-compliant.",
+  "version": "3.0.3",
   "author": "Ivan Nikulin <ifaaan@gmail.com> (https://github.com/inikulin)",
-  "contributors": [
-    "Alan Clarke (https://github.com/alanclarke)",
-    "Evan You (http://evanyou.me)",
-    "Saksham Aggarwal <s.agg2021@gmail.com>",
-    "Sebastian Mayr <sebmaster16@gmail.com> (http://blog.smayr.name)",
-    "Sean Lang <slang800@gmail.com> (http://slang.cx)"
-  ],
+  "contributors": "https://github.com/inikulin/parse5/graphs/contributors",
   "homepage": "https://github.com/inikulin/parse5",
   "devDependencies": {
     "del": "^2.0.2",
     "gulp": "^3.9.0",
     "gulp-benchmark": "^1.1.1",
-    "gulp-concat": "^2.6.0",
     "gulp-download": "0.0.1",
     "gulp-eslint": "^3.0.1",
-    "gulp-insert": "^0.5.0",
     "gulp-install": "^0.6.0",
-    "gulp-jsdoc-to-markdown": "^1.1.1",
     "gulp-mocha": "^2.1.3",
     "gulp-rename": "^1.2.2",
+    "gulp-typedoc": "^2.0.0",
+    "gulp-typescript": "^3.1.2",
     "publish-please": "^2.2.0",
-    "through2": "^2.0.0"
+    "through2": "^2.0.0",
+    "typedoc": "^0.5.1",
+    "typescript": "^2.0.6"
   },
   "keywords": [
     "html",
@@ -49,6 +44,7 @@
   ],
   "license": "MIT",
   "main": "./lib/index.js",
+  "types": "./lib/index.d.ts",
   "repository": {
     "type": "git",
     "url": "git://github.com/inikulin/parse5.git"
@@ -60,5 +56,8 @@
   },
   "files": [
     "lib"
-  ]
+  ],
+  "dependencies": {
+    "@types/node": "*"
+  }
 }

package/lib/location_info/parser_mixin.js

@@ -1,217 +0,0 @@
-'use strict';
-
-var OpenElementStack = require('../parser/open_element_stack'),
-    Tokenizer = require('../tokenizer'),
-    HTML = require('../common/html');
-
-
-//Aliases
-var $ = HTML.TAG_NAMES;
-
-
-function setEndLocation(element, closingToken, treeAdapter) {
-    var loc = element.__location;
-
-    if (!loc)
-        return;
-
-    /**
-     * @typedef {Object} ElementLocationInfo
-     * @extends StartTagLocationInfo
-     *
-     * @property {StartTagLocationInfo} startTag - Element's start tag location info.
-     * @property {LocationInfo} endTag - Element's end tag location info.
-     */
-    if (!loc.startTag) {
-        loc.startTag = {
-            line: loc.line,
-            col: loc.col,
-            startOffset: loc.startOffset,
-            endOffset: loc.endOffset
-        };
-        if (loc.attrs)
-            loc.startTag.attrs = loc.attrs;
-    }
-
-    if (closingToken.location) {
-        var ctLocation = closingToken.location,
-            tn = treeAdapter.getTagName(element),
-        // NOTE: For cases like <p> <p> </p> - First 'p' closes without a closing tag and
-        // for cases like <td> <p> </td> - 'p' closes without a closing tag
-            isClosingEndTag = closingToken.type === Tokenizer.END_TAG_TOKEN &&
-                              tn === closingToken.tagName;
-
-        if (isClosingEndTag) {
-            loc.endTag = {
-                line: ctLocation.line,
-                col: ctLocation.col,
-                startOffset: ctLocation.startOffset,
-                endOffset: ctLocation.endOffset
-            };
-        }
-
-        if (isClosingEndTag)
-            loc.endOffset = ctLocation.endOffset;
-        else
-            loc.endOffset = ctLocation.startOffset;
-    }
-}
-
-
-exports.assign = function (parser) {
-    //NOTE: obtain Parser proto this way to avoid module circular references
-    var parserProto = Object.getPrototypeOf(parser),
-        treeAdapter = parser.treeAdapter,
-        attachableElementLocation = null,
-        lastFosterParentingLocation = null,
-        currentToken = null;
-
-
-    //NOTE: patch _bootstrap method
-    parser._bootstrap = function (document, fragmentContext) {
-        parserProto._bootstrap.call(this, document, fragmentContext);
-
-        attachableElementLocation = null;
-        lastFosterParentingLocation = null;
-        currentToken = null;
-
-        //OpenElementStack
-        parser.openElements.pop = function () {
-            setEndLocation(this.current, currentToken, treeAdapter);
-            OpenElementStack.prototype.pop.call(this);
-        };
-
-        parser.openElements.popAllUpToHtmlElement = function () {
-            for (var i = this.stackTop; i > 0; i--)
-                setEndLocation(this.items[i], currentToken, treeAdapter);
-
-            OpenElementStack.prototype.popAllUpToHtmlElement.call(this);
-        };
-
-        parser.openElements.remove = function (element) {
-            setEndLocation(element, currentToken, treeAdapter);
-            OpenElementStack.prototype.remove.call(this, element);
-        };
-    };
-
-
-    //Token processing
-    parser._processTokenInForeignContent = function (token) {
-        currentToken = token;
-        parserProto._processTokenInForeignContent.call(this, token);
-    };
-
-    parser._processToken = function (token) {
-        currentToken = token;
-        parserProto._processToken.call(this, token);
-
-        //NOTE: <body> and <html> are never popped from the stack, so we need to updated
-        //their end location explicitly.
-        if (token.type === Tokenizer.END_TAG_TOKEN &&
-            (token.tagName === $.HTML ||
-             token.tagName === $.BODY && this.openElements.hasInScope($.BODY))) {
-            for (var i = this.openElements.stackTop; i >= 0; i--) {
-                var element = this.openElements.items[i];
-
-                if (this.treeAdapter.getTagName(element) === token.tagName) {
-                    setEndLocation(element, token, treeAdapter);
-                    break;
-                }
-            }
-        }
-    };
-
-
-    //Doctype
-    parser._setDocumentType = function (token) {
-        parserProto._setDocumentType.call(this, token);
-
-        var documentChildren = this.treeAdapter.getChildNodes(this.document),
-            cnLength = documentChildren.length;
-
-        for (var i = 0; i < cnLength; i++) {
-            var node = documentChildren[i];
-
-            if (this.treeAdapter.isDocumentTypeNode(node)) {
-                node.__location = token.location;
-                break;
-            }
-        }
-    };
-
-
-    //Elements
-    parser._attachElementToTree = function (element) {
-        //NOTE: _attachElementToTree is called from _appendElement, _insertElement and _insertTemplate methods.
-        //So we will use token location stored in this methods for the element.
-        element.__location = attachableElementLocation || null;
-        attachableElementLocation = null;
-        parserProto._attachElementToTree.call(this, element);
-    };
-
-    parser._appendElement = function (token, namespaceURI) {
-        attachableElementLocation = token.location;
-        parserProto._appendElement.call(this, token, namespaceURI);
-    };
-
-    parser._insertElement = function (token, namespaceURI) {
-        attachableElementLocation = token.location;
-        parserProto._insertElement.call(this, token, namespaceURI);
-    };
-
-    parser._insertTemplate = function (token) {
-        attachableElementLocation = token.location;
-        parserProto._insertTemplate.call(this, token);
-
-        var tmplContent = this.treeAdapter.getTemplateContent(this.openElements.current);
-
-        tmplContent.__location = null;
-    };
-
-    parser._insertFakeRootElement = function () {
-        parserProto._insertFakeRootElement.call(this);
-        this.openElements.current.__location = null;
-    };
-
-
-    //Comments
-    parser._appendCommentNode = function (token, parent) {
-        parserProto._appendCommentNode.call(this, token, parent);
-
-        var children = this.treeAdapter.getChildNodes(parent),
-            commentNode = children[children.length - 1];
-
-        commentNode.__location = token.location;
-    };
-
-
-    //Text
-    parser._findFosterParentingLocation = function () {
-        //NOTE: store last foster parenting location, so we will be able to find inserted text
-        //in case of foster parenting
-        lastFosterParentingLocation = parserProto._findFosterParentingLocation.call(this);
-        return lastFosterParentingLocation;
-    };
-
-    parser._insertCharacters = function (token) {
-        parserProto._insertCharacters.call(this, token);
-
-        var hasFosterParent = this._shouldFosterParentOnInsertion(),
-            parent = hasFosterParent && lastFosterParentingLocation.parent ||
-                     this.openElements.currentTmplContent ||
-                     this.openElements.current,
-            siblings = this.treeAdapter.getChildNodes(parent),
-            textNodeIdx = hasFosterParent && lastFosterParentingLocation.beforeElement ?
-            siblings.indexOf(lastFosterParentingLocation.beforeElement) - 1 :
-            siblings.length - 1,
-            textNode = siblings[textNodeIdx];
-
-        //NOTE: if we have location assigned by another token, then just update end position
-        if (textNode.__location)
-            textNode.__location.endOffset = token.location.endOffset;
-
-        else
-            textNode.__location = token.location;
-    };
-};
-

package/lib/location_info/tokenizer_mixin.js

@@ -1,169 +0,0 @@
-'use strict';
-
-var UNICODE = require('../common/unicode');
-
-//Aliases
-var $ = UNICODE.CODE_POINTS;
-
-
-exports.assign = function (tokenizer) {
-    //NOTE: obtain Tokenizer proto this way to avoid module circular references
-    var tokenizerProto = Object.getPrototypeOf(tokenizer),
-        tokenStartOffset = -1,
-        tokenCol = -1,
-        tokenLine = 1,
-        isEol = false,
-        lineStartPosStack = [0],
-        lineStartPos = 0,
-        col = -1,
-        line = 1;
-
-    function attachLocationInfo(token) {
-        /**
-         * @typedef {Object} LocationInfo
-         *
-         * @property {Number} line - One-based line index
-         * @property {Number} col - One-based column index
-         * @property {Number} startOffset - Zero-based first character index
-         * @property {Number} endOffset - Zero-based last character index
-         */
-        token.location = {
-            line: tokenLine,
-            col: tokenCol,
-            startOffset: tokenStartOffset,
-            endOffset: -1
-        };
-    }
-
-    //NOTE: patch consumption method to track line/col information
-    tokenizer._consume = function () {
-        var cp = tokenizerProto._consume.call(this);
-
-        //NOTE: LF should be in the last column of the line
-        if (isEol) {
-            isEol = false;
-            line++;
-            lineStartPosStack.push(this.preprocessor.sourcePos);
-            lineStartPos = this.preprocessor.sourcePos;
-        }
-
-        if (cp === $.LINE_FEED)
-            isEol = true;
-
-        col = this.preprocessor.sourcePos - lineStartPos + 1;
-
-        return cp;
-    };
-
-    tokenizer._unconsume = function () {
-        tokenizerProto._unconsume.call(this);
-        isEol = false;
-
-        while (lineStartPos > this.preprocessor.sourcePos && lineStartPosStack.length > 1) {
-            lineStartPos = lineStartPosStack.pop();
-            line--;
-        }
-
-        col = this.preprocessor.sourcePos - lineStartPos + 1;
-    };
-
-    //NOTE: patch token creation methods and attach location objects
-    tokenizer._createStartTagToken = function () {
-        tokenizerProto._createStartTagToken.call(this);
-        attachLocationInfo(this.currentToken);
-    };
-
-    tokenizer._createEndTagToken = function () {
-        tokenizerProto._createEndTagToken.call(this);
-        attachLocationInfo(this.currentToken);
-    };
-
-    tokenizer._createCommentToken = function () {
-        tokenizerProto._createCommentToken.call(this);
-        attachLocationInfo(this.currentToken);
-    };
-
-    tokenizer._createDoctypeToken = function (initialName) {
-        tokenizerProto._createDoctypeToken.call(this, initialName);
-        attachLocationInfo(this.currentToken);
-    };
-
-    tokenizer._createCharacterToken = function (type, ch) {
-        tokenizerProto._createCharacterToken.call(this, type, ch);
-        attachLocationInfo(this.currentCharacterToken);
-    };
-
-    tokenizer._createAttr = function (attrNameFirstCh) {
-        tokenizerProto._createAttr.call(this, attrNameFirstCh);
-        this.currentAttrLocation = {
-            line: line,
-            col: col,
-            startOffset: this.preprocessor.sourcePos,
-            endOffset: -1
-        };
-    };
-
-    tokenizer._leaveAttrName = function (toState) {
-        tokenizerProto._leaveAttrName.call(this, toState);
-        this._attachCurrentAttrLocationInfo();
-    };
-
-    tokenizer._leaveAttrValue = function (toState) {
-        tokenizerProto._leaveAttrValue.call(this, toState);
-        this._attachCurrentAttrLocationInfo();
-    };
-
-    tokenizer._attachCurrentAttrLocationInfo = function () {
-        this.currentAttrLocation.endOffset = this.preprocessor.sourcePos;
-
-        if (!this.currentToken.location.attrs)
-            this.currentToken.location.attrs = {};
-
-        /**
-         * @typedef {Object} StartTagLocationInfo
-         * @extends LocationInfo
-         *
-         * @property {Dictionary<String, LocationInfo>} attrs - Start tag attributes' location info.
-         */
-        this.currentToken.location.attrs[this.currentAttr.name] = this.currentAttrLocation;
-    };
-
-    //NOTE: patch token emission methods to determine end location
-    tokenizer._emitCurrentToken = function () {
-        //NOTE: if we have pending character token make it's end location equal to the
-        //current token's start location.
-        if (this.currentCharacterToken)
-            this.currentCharacterToken.location.endOffset = this.currentToken.location.startOffset;
-
-        this.currentToken.location.endOffset = this.preprocessor.sourcePos + 1;
-        tokenizerProto._emitCurrentToken.call(this);
-    };
-
-    tokenizer._emitCurrentCharacterToken = function () {
-        //NOTE: if we have character token and it's location wasn't set in the _emitCurrentToken(),
-        //then set it's location at the current preprocessor position.
-        //We don't need to increment preprocessor position, since character token
-        //emission is always forced by the start of the next character token here.
-        //So, we already have advanced position.
-        if (this.currentCharacterToken && this.currentCharacterToken.location.endOffset === -1)
-            this.currentCharacterToken.location.endOffset = this.preprocessor.sourcePos;
-
-        tokenizerProto._emitCurrentCharacterToken.call(this);
-    };
-
-    //NOTE: patch initial states for each mode to obtain token start position
-    Object.keys(tokenizerProto.MODE)
-
-        .map(function (modeName) {
-            return tokenizerProto.MODE[modeName];
-        })
-
-        .forEach(function (state) {
-            tokenizer[state] = function (cp) {
-                tokenStartOffset = this.preprocessor.sourcePos;
-                tokenLine = line;
-                tokenCol = col;
-                tokenizerProto[state].call(this, cp);
-            };
-        });
-};

package/lib/parser/stream.js

@@ -1,140 +0,0 @@
-'use strict';
-
-var WritableStream = require('stream').Writable,
-    inherits = require('util').inherits,
-    Parser = require('./index');
-
-/**
- * Streaming HTML parser with scripting support.
- * A [writable stream]{@link https://nodejs.org/api/stream.html#stream_class_stream_writable}.
- * @class ParserStream
- * @memberof parse5
- * @instance
- * @extends stream.Writable
- * @param {ParserOptions} options - Parsing options.
- * @example
- * var parse5 = require('parse5');
- * var http = require('http');
- *
- * // Fetch the google.com content and obtain it's <body> node
- * http.get('http://google.com', function(res) {
- *  var parser = new parse5.ParserStream();
- *
- *  parser.on('finish', function() {
- *      var body = parser.document.childNodes[0].childNodes[1];
- *  });
- *
- *  res.pipe(parser);
- * });
- */
-var ParserStream = module.exports = function (options) {
-    WritableStream.call(this);
-
-    this.parser = new Parser(options);
-
-    this.lastChunkWritten = false;
-    this.writeCallback = null;
-    this.pausedByScript = false;
-
-    /**
-     * The resulting document node.
-     * @member {ASTNode<document>} document
-     * @memberof parse5#ParserStream
-     * @instance
-     */
-    this.document = this.parser.treeAdapter.createDocument();
-
-    this.pendingHtmlInsertions = [];
-
-    this._resume = this._resume.bind(this);
-    this._documentWrite = this._documentWrite.bind(this);
-    this._scriptHandler = this._scriptHandler.bind(this);
-
-    this.parser._bootstrap(this.document, null);
-};
-
-inherits(ParserStream, WritableStream);
-
-//WritableStream implementation
-ParserStream.prototype._write = function (chunk, encoding, callback) {
-    this.writeCallback = callback;
-    this.parser.tokenizer.write(chunk.toString('utf8'), this.lastChunkWritten);
-    this._runParsingLoop();
-};
-
-ParserStream.prototype.end = function (chunk, encoding, callback) {
-    this.lastChunkWritten = true;
-    WritableStream.prototype.end.call(this, chunk, encoding, callback);
-};
-
-//Scriptable parser implementation
-ParserStream.prototype._runParsingLoop = function () {
-    this.parser._runParsingLoop(this.writeCallback, this._scriptHandler);
-};
-
-ParserStream.prototype._resume = function () {
-    if (!this.pausedByScript)
-        throw new Error('Parser was already resumed');
-
-    while (this.pendingHtmlInsertions.length) {
-        var html = this.pendingHtmlInsertions.pop();
-
-        this.parser.tokenizer.insertHtmlAtCurrentPos(html);
-    }
-
-    this.pausedByScript = false;
-
-    //NOTE: keep parsing if we don't wait for the next input chunk
-    if (this.parser.tokenizer.active)
-        this._runParsingLoop();
-};
-
-ParserStream.prototype._documentWrite = function (html) {
-    if (!this.parser.stopped)
-        this.pendingHtmlInsertions.push(html);
-};
-
-ParserStream.prototype._scriptHandler = function (scriptElement) {
-    if (this.listeners('script').length) {
-        this.pausedByScript = true;
-
-        /**
-         * Raised then parser encounters a `<script>` element.
-         * If this event has listeners, parsing will be suspended once it is emitted.
-         * So, if `<script>` has the `src` attribute, you can fetch it, execute and then resume parsing just like browsers do.
-         * @event script
-         * @memberof parse5#ParserStream
-         * @instance
-         * @type {Function}
-         * @param {ASTNode} scriptElement - The script element that caused the event.
-         * @param {Function} documentWrite(html) - Write additional `html` at the current parsing position.
-         *  Suitable for implementing the DOM `document.write` and `document.writeln` methods.
-         * @param {Function} resume - Resumes parsing.
-         * @example
-         * var parse = require('parse5');
-         * var http = require('http');
-         *
-         * var parser = new parse5.ParserStream();
-         *
-         * parser.on('script', function(scriptElement, documentWrite, resume) {
-         *   var src = parse5.treeAdapters.default.getAttrList(scriptElement)[0].value;
-         *
-         *   http.get(src, function(res) {
-         *      // Fetch the script content, execute it with DOM built around `parser.document` and
-         *      // `document.write` implemented using `documentWrite`.
-         *      ...
-         *      // Then resume parsing.
-         *      resume();
-         *   });
-         * });
-         *
-         * parser.end('<script src="example.com/script.js"></script>');
-         */
-
-
-        this.emit('script', scriptElement, this._documentWrite, this._resume);
-    }
-    else
-        this._runParsingLoop();
-};
-